Identification of genes that confer susceptibility to insulin-dependent diabetes mellitus (IDDM) will provide: 1) important new information about the pathogenesis of the disease and 2) a means for identifying individuals who may benefit from the protective therapies that are currently under investigation. Herein, we propose a continuation of our studies aimed at identifying susceptibility genes involved in IDDM, an autoimmune disease in which the insulin-producing, pancreatic beta cells are destroyed. We have already analyzed 104 affected sibpair (ASP) families and identified 5 genomic intervals that are linked to IDDM, in addition to the long recognized IDDM1 (HLA) and IDDM2 (INS region). IDDM3 has been mapped to 15q26, IDDM4 to 11q13, IDDM5 to 6q24, IDDM7 to 2q31- q33. These intervals were also reported by other investigators. We were the first group to map IDDM8 to a 21cM genomic region on 6q25-q27. In addition, suggestive evidence of linkage has been reported for more than 10 other genomic regions. Since IDDM is genetically heterogeneous and subject to environmental influences, most of the linkages still require confirmation in additional data sets. Thus, we will exploit our unique computerized network of IDDM families to expand by 150, our well established collection of ASP families. We will then use our newly ascertained family set to confirm all putative linkage intervals. As in the past, our new families will be made available to other interested investigators worldwide under the auspices of the human biological data interchange (HBDI). Outside of the HLA complex, the evidence of linkage for IDDM8 (MLS =3.4) and its impact on disease (lambdas = 1.8) are the strongest among all susceptibility intervals identified in our data set. We therefore propose to carry out fine mapping for this linkage interval. ASP analysis in 500 families will first be used to narrow the IDDM8 interval within 3-ScM. Linkage disequilibrium analysis will then be performed using the ASP and simplex families as well as unrelated patient and control populations from diverse ethnic groups (Caucasians, Chinese and African Americans). The ultimate goal of disequilibrium analysis is to detect marker-disease association and to define the genomic region that is significantly associated with the disease. Candidate genes within a genetic interval merit special attention. Within the IDDM8 interval, 2 such loci have been identified, SOD2 and IGF2R. Our preliminary observations indicate that the linkage evidence for IDDM8 is consistent with genomic imprinting, a pattern similar to that already reported for IGF2R. Provocatively, a paternally-imprinted gene responsible for beta cell disfunction in transient neonatal diabetes mellitus has also been mapped to chromosome 6. This proposal will provide important information for the ultimate identification of IDDM genes by positional cloning techniques.